8SEP

France's G7 text drops cloud sovereignty

3 min read

15:06UTC

The G7 Digital Ministerial Declaration, signed at Bercy in Paris on 29 May under French chairmanship, contained no mention of cloud sovereignty, CAIDA, or restrictions on US cloud providers.

← Back to Mistral closes EUR 3bn; Luxembourg buys in Jump to analysis ↓

TechnologyDeveloping

Key takeaway

France chaired the G7 and left cloud sovereignty out of the text, exposing CAIDA's lack of international support.

The G7 Digital and Technology Ministerial Declaration, signed at Bercy in Paris on Friday 29 May under French chairmanship, contained no mention of cloud sovereignty, CAIDA, or restrictions on US cloud providers ¹. The G7 is the bloc of seven leading advanced economies; its digital ministers had gathered to agree shared priorities. The four they signed were AI security, AI openness for smaller firms, digital-sector resilience, and child safety online. The text held to the Hiroshima AI Process, the G7's 2023 AI-safety framework, rather than to France's own "Cloud au Centre" doctrine.

The signed communique confirmed what the ministerial agenda had already foreshadowed . The two frames in play do not combine: the Hiroshima AI Process treats AI as a safety problem to be governed cooperatively, while the sovereignty frame treats US cloud dominance as a strategic dependency to be reduced. A room containing the United States and Japan would accept only the first. France, chairing without an adopted package to table , had no instrument to defend and chose the text its partners could sign.

That choice is the measure of CAIDA's international isolation. The country leading the sovereignty argument inside the EU would not, or could not, carry it into a venue where Washington could object. The law's strongest advocate declined to defend it where defending it had a cost, which leaves Brussels adopting an instrument no G7 communique endorses.

Deep Analysis

In plain English

The G7 is a group of seven major democracies, including France, Germany, the US, the UK and Japan, that meet regularly to coordinate policy. In late May 2026, France chaired a G7 meeting specifically on digital technology in Paris. France had been expected to use the meeting to push other G7 members toward its vision of digital sovereignty, which involves restricting American cloud companies from handling sensitive data. Instead, the final agreed statement said nothing about cloud sovereignty at all. The four topics everyone agreed on were AI safety, helping smaller businesses use AI, protecting children online, and making digital systems more resilient against outages and attacks.

Deep Analysis

Root Causes

The Bercy outcome has two structural causes unrelated to French diplomatic skill. First, the G7 digital ministerial has no enforcement mechanism: it produces declarations, not binding instruments. This means the only value of sovereignty language in the text would be signalling and precedent, and France judged the signalling value insufficient to fight for against US opposition.

Second, France chaired without an adopted Tech Sovereignty Package to put on the table. CAIDA's fourth consecutive slip meant France had no legislative fact to anchor sovereignty language to. A G7 presidency cannot credibly demand international recognition of a doctrine its own Commission has failed to legislate in four attempts. France removed cloud sovereignty from the agenda because CAIDA's fourth slip left it without a legislative fact to anchor any sovereignty demand to.

What could happen next?

Consequence
The Bercy precedent reduces the probability that a future G7 digital ministerial will include cloud-sovereignty language, even after CAIDA adopts, because the US will cite the Bercy silence as the baseline.
Medium term · Reported
Risk
A French presidency that could not secure sovereignty language at its own G7 ministerial weakens the diplomatic credibility of EU arguments at the WTO and in bilateral EU-US digital trade talks.
Short term · Reported
Meaning
The Hiroshima AI Process framework, which the Bercy declaration reaffirmed, was authored under Japanese presidency and does not include EU sovereignty positions; Bercy's reaffirmation locks the G7 to that baseline for another cycle.
Long term · Assessed

Source Landscape

This story draws on neutral-leaning sources

Primary parallel: The 2016 G7 Taormina summit under Italian presidency produced a similar dynamic on data localisation. Italy had pushed for language acknowledging member states' rights to restrict cross-border data flows on security grounds; the final communique omitted any such language after US objections. The pattern is consistent across G7 digital ministerials: sovereignty language that constrains US platforms has never survived a jointly signed text.

Counter-parallel: The 2019 Osaka G20 Leaders' Declaration achieved the 'Osaka Track' on data free flow with trust, signed by all members including the EU. Japan structured the initiative so that both sides could claim victory: the US got the free-flow language, the EU got the 'with trust' qualifier that preserved GDPR compatibility.

France's failure to apply a similar linguistic bridge at Bercy suggests the French presidency did not prioritise face-saving language engineering over substantive sovereignty text.

Consensus view: Sciences Po Paris political scientist Cornelia Woll and the European Council on Foreign Relations' (ECFR) tech programme lead Janka Oertel both assess that the Bercy communique's silence on cloud sovereignty and CAIDA reflects a French presidency trade-off: France needed G7 AI consensus on child safety and SME access, areas where Washington agreed, and sacrificed the sovereignty language to get the declaration signed.

A communique with reservations would have been worse for France's G7 ambitions than a clean text that simply omits the contested material.

Counter-view: The Cato Institute's technology director Matthew Feeney reads the outcome differently: France's own Cloud au centre doctrine was never compatible with a US-co-signed G7 declaration, and the Bercy text reveals that French sovereignty positions are primarily domestic rhetoric that Paris does not actually press at multilateral level. CAIDA's concurrent delay gives Washington no reason to take French sovereignty demands seriously as a negotiating position.

Key tension: Whether the G7 is a venue where European digital sovereignty positions can be advanced at all, given that the US, Canada and Japan together hold a structural veto over any tech-policy language that constrains US-platform incumbents.

First Reported In

Update #7 · Sovereignty arrives, minus Brussels

G7 France· 3 Jun 2026

Read original →

Causes and effects

Caused by

France chairs G7 with nothing to table

The G7 Bercy agenda had already omitted cloud sovereignty; the signed 29 May declaration confirmed that omission as the final agreed text.

Occurred 29 May 2026

Read story →

CAIDA due before College, scope cut

France chaired the G7 ministerial without an adopted package to table, which directly explains why it produced a communique that did not mention CAIDA or cloud sovereignty.

Occurred 3 Jun 2026

Read story →

This Event

France's G7 text drops cloud sovereignty

France led the meeting and produced a text closer to Washington's framing than its own, the clearest sign CAIDA has no international backing.

Different Perspectives

Poland

Poland leads a self-announced AI Gigafactory consortium with a EUR 100 million phase-one commitment, matched by Czechia and joined by Hungary at EUR 25 million. EuroHPC has confirmed no consortium for the call closing 12 November, so the bloc exists only in national announcements so far.

UK Government (DCMS)

Secretary of State Lisa Nandy told the Commons on 3 September she has inherited the sovereign AI brief from Liz Kendall, but gave no assessment, figure or date on frontier-model access. It is the first public claim of ownership since DSIT's abolition, without the substance the committee asked for.

ASML

ASML CEO Christophe Fouquet credited Intel Foundry, not a European fab, with shipping the first high-volume logic product made on High-NA EUV, the tool only ASML builds. Europe holds the chokepoint tool; the company that spent it into volume production first is American.

Luxembourg

Luxembourg joined Mistral's Series D as a new investor on 8 September, the same government that co-funds EuroHPC's MeluXina-AI supercomputer. One small member state now funds the sovereign compute Mistral may need and holds equity in the company using it.

Samsung Electronics

Samsung led Mistral's Series D on 8 September and the same day expanded its ASML collaboration on next-generation lithography. One Korean company now sits atop Europe's largest AI funding round and inside its most sensitive chip-tooling relationship.

Mistral AI

Mistral closed a EUR 3 billion Series D above a EUR 21 billion valuation on 8 September, with Samsung leading and Luxembourg joining as a new state investor. The company markets itself as Europe's non-American AI option even as the round's lead capital comes from South Korea and two US private-equity firms.